Over the years, various microelectromechanical systems ("MEMS") have arisen which require the necessity to sense temperature, pressure, strain, acceleration, rotation, infrared radiation, chemical properties of liquids and gases, and other physical inputs. Accordingly, various types of microsensors have been developed which receive analog and digital electrical inputs and also sense or measure these other physical inputs, e.g., acceleration, pressure, temperature, strain.
Integrated circuits are widely used in many of these MEMS or electronic applications. Various integrated circuit manufacturing processes, e.g., very large scale integrated ("VLSI") are also widely known and provide various advantages. The complimentary metal oxide semiconductor ("CMOS") manufacturing technology, for example, generally provides a low power dissipation advantage over known metal oxide semiconductor ("MOS") processes. Microsensor manufacturing which is compatible with known integrated circuit manufacturing processes, however, can be quite complicated, especially because of a need for integrating various types of structures at relatively low cost.
Examples of applications for microsensors for acceleration or accelerometers include air bag systems, anti-lock braking systems, and ride suspension systems for automobiles and in-flight aircraft monitoring systems for aircraft. Each of these applications requires small, inexpensive, and reliable acceleration devices.
Many of the known accelerometers for these applications, for example, are analog and measure or sense an electrical current that varies with frequency or amplitude of acceleration. In other words, in essence, many of these sensors convert mechanical parameters to other energy domains and then sense or measure directly. For sensors using direct sensing, the parameters are conventionally related to strain, stress, or displacement. The principles conventionally used to measure or sense strain are piezoelectricity, piezoresistivity, and capacitive or inductive impedance.
The measurement of piezoelectric effects, however, often requires a high input impedance amplifier to measure the surface charges or voltages generated by the stress or the strain. These types of sensors can be expensive and are often not readily acceptable for high density integrated circuit technology and various integrated circuit manufacturing technology.
The measurement of piezoresistivity in conductors and semiconductors conventionally involves the strain on the crystal structure deforming the energy band structure and, thus, changing the mobility and carrier density that changes the resistivity or orientation. These type of sensors, however, are also like piezoelectric sensors in that these sensors can be expensive to manufacture and often may not be very stable for acceleration applications.
Capacitive or inductive impedances can also be used to measure acceleration. Examples of such sensors can be seen in U.S. Pat. No. 5,417,312 by Tsuchitani et al. titled "Semiconductor Acceleration Sensor and Vehicle Control System Using The Same," U.S. Pat. No. 5,506,454 by Hanzawa et al. titled "System And Method For Diagnosing Characteristics Of Acceleration Sensor," U.S. Pat. No. 5,610,335 by Shaw et al. titled "Microelectromechanical Lateral Accelerometer," and U.S. Pat. No. 5,659,195 by Kaiser et al. titled "CMOS Integrated Microsensor With A Precision Measurement Circuit." Capacitive devices integrate the change of elementary capacitive areas while piezoresistive devices take the difference of the resistance changes of bridge arms. Accordingly, capacitive sensors are generally less sensitive to the sideways or indirect forces and are generally more stable. Capacitive sensors, however, conventionally require a capacitance-to-voltage converter on or near the chip to avoid the effects of stray capacitances which can complicate the associated circuitry. The measurement circuitry for these types of sensors is also required to be stable and have low noise.
Additionally, some accelerometers provide a digital output by using a "spring" that either makes or breaks an electrical contact in response to acceleration. Some of these spring elements, for example, may provide a series of sensing elements having incrementally higher response thresholds which make electrical contact when the threshold is reached. These "spring" accelerometers, however, are relatively large in size as compared to VLSI circuitry and can be quite difficult to make compatible with current integrated circuit manufacturing processes.
Yet further types of microsensors which provide a digital output for detecting translational or rotational acceleration are also known. An example of such a microsersor can be seen in U.S. Pat. No. 5,610,337 by Nelson titled "Method of Measuring The Amplitude And Frequency Of An Acceleration." One type of accelerometer uses a sensing element which has some sort of pivotally mounted tilting beam (see FIG. 1B therein). The pivotally mounted tilting beam includes a hinge portion, a rigid connection member connected to the hinge portion, and a pair of respective end portions, e.g., a proof mass, connected to the rigid connection member. The end portions rotate clockwise and counter clockwise during rotational, e.g., horizontal, movement. Only one of the end portions contacts a contact electrode which responsively stores the contact signal to indicate that the movement was in the one direction. In essence, this sensing element provides three-states, namely tilted contact in one direction, tilted contact in the other direction, or untilted or neutral. Such a sensing element, however, requires a reset pulse or a reset position which needs to be activated by an external reset activation source.
Another type of accelerometer which provides a digital output is also illustrated in this patent (see FIGS. 2-4). This sensing element provides a cantilever beam type arrangement that includes a thick beam, a thinner portion of flexible material connected to and extending outwardly from the thick beam and defining a hinge, and a thicker proof mass or end portion connected to and extending outwardly from the hinge. This arrangement of a cantilever beam has problems with "stuck on" conditions which also require complex reset structures and conditions. This arrangement also may include small critical dimension which can make manufacturing of such a device difficult and expensive with known integrated circuit manufacturing processes such as CMOS technology.